9307 ‎(2)‎

• UCREFRP Catalog Number: 9307
• Author: Germaine, S. S.
• Title: Relationships of Birds, Lizards, and Nocturnal Rodents to Their Habitat in the Greater Tucson Area, Arizona.
• USFW Year: 1995.
• USFW - Doc Type: Phoenix, AZ.
• Copyright Material: NO

RELATIONSHIPS OF BIRDS, LIZARDS, AND NOCTURNAL RODENTS TO THEIR HABITAT IN TUCSON, ARIZONA
birds. However, only 306 plots were visited, and
1 visit per plot was made.
Lizards. I surveyed lizards in a systematic
subset of 130 census plots from July 12 through
September 14, 1994. On each transect I censused
plot nos. 1, 4, 7, and 10, except 2 plots on which
new home construction had begun since the
breeding bird censusing ended. censusing was
begun each morning 1 hour after sunrise and
ended 5 hours after sunrise. To account for
temperature related differences in activity peaks
among lizard species, I divided each morning's
census period into 2 2-hour time blocks. Each
census plot was visited once per time block. Time
blocks were censused back-to-back, and I censused
4 plots in each block each day.
At each plot a 200-m transect centered on the
plot center was walked at a steady pace (50 m/5
to 6 minutes). At plots occurring on roads, 100-m
transects were walked on each side of the road
and summed. At non-road plots, a straight 200-m
transect was walked. This transect was centered
on the plot center and followed the existing trail.
Only lizards within IS m of the transect were
included. Once a lizard was spotted, time spent
identifying it (if necessary) was not included in the
elapsed census time. Interruptions were treated in
the same manner as in bird censusing. Transect
route road surfaces were excluded from surveys,
and an equal (non) likelihood of lizards crossing
roads and being re-surveyed was assumed at each
transect.
Nocturnal Rodents. I collected abundance data
for nocturnal rodents from 8 50-trap lines of snap
traps from October 3-5, 1994. I limited the
nocturnal rodent component of this study to a
comparison of the assemblages present at 2 pre-
defined levels of housing density. I placed 2 trap
lines in low density housing (0.5 houses/ha) and 2
in high density housing (7.5 houses/ha). Each
housing treatment was adjacent to a >_ 10 ha
control block of undeveloped habitat which
contained native vegetation (0 house density sites).
For each of the 4 trap lines placed within
residential areas, I placed a control trap line near
the center of the native habitat block. This
arrangement resulted in 4 treatments: 1) low
density, 2) low-density control, 3) high density,
and 4) high-density control. Paired study sites
were similar with respect to vegetation types,
proximity to washes, and elevation. Native
habitat fragments were no more than twice as
long as wide to minimize edge effects, and I
located all of the sampling sites in an area
undergoing rapid residential development.
I placed 2 traps every 10 m, totalling 250 m
trapping distance per line. In low-density sites,
traps were placed 2 m from the edge of roads. In
high-density sites, traps were placed along the
edges of alley-ways in areas containing few solid
block walls which might restrict small mammal
movement. Traps in control sites were set at the
same spacing as used at residential sites. Mouse
traps (Museum Special and Victor) were set at a
3:1 ratio to rat traps (Victor). Traps were baited
with a rolled oat/peanut butter mixture, set at
dusk of each night of sampling, and collected
shortly after dawn the following day.
Land Cover Quantification
The focus of habitat quantification in this
study was: 1) to measure the percent of land cover
that was altered from its native state, and 2) to
identify the degree to which it was altered. To
describe the existing land cover types and
arrangements, I measured 33 habitat variables
from aerial photographs and adjusted them with
field reconnaissance (Table 1). I selected or
modified several variables from Stenberg (1988)
and Shaw et al. (1993) to be more compatible with
other studies. Also, several variables were
measured only for descriptive purposes and were
not used in statistical analyses.
Primary land cover type was recorded for
each census plot as either lower Sonoran (Larrea
tridentata dominant), upper Sonoran (Cercidium
spp.-Carnegiea gigantea dominant), or non-native
(no land left in a native vegetative state, all
vegetative cover consisting of ornamental
plantings or weedy growth).
I determined percent area in each land cover
type from measurements of aerial photographs
and from field verification of land cover category
assignments. I centered an acetate circle
representing a 100-m radius (3.1 ha) circular plot
on each census plot. The acetate overlay
contained a 3.6-m dot grid on which I traced
polygons delimiting each land cover type present.
A 100-m radius plot was chosen to better estimate
the actual value for the habitat cover types at each
plot, because wildlife were not restricted by the
boundaries of the plots. I then calculated the
percent area of each polygon and summed these
within each cover type.
ARIZONA GAME fi FISH DEPARTMENT, TECH. REP. 20 STEPHEN S. GERMAINE 1995